Video-to-music generation on V2M-bench (test)
0Fréchet Audio Distance (FAD)GT
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| GT2024.06 | 0 | 0 | 0 | 1.167 | 1 | 0.241 | |
| Audio-OmniModel Type=Unified Models2026.04 | 1.58 | — | — | — | — | — | |
| AudioXModel Type=Specialized Models2026.04 | 2.12 | — | — | — | — | — | |
| VidMuse2024.06 | 2.459 | 0.734 | 29.946 | 1.25 | 0.73 | 0.202 | |
| VidMuseModel Type=Specialized Models2026.04 | 2.46 | — | — | — | — | — | |
| Caption2Music2024.06 | 2.485 | 1.081 | 40.199 | 0.378 | 0.486 | 0.191 | |
| M2UGen*variant=asterisk variant2024.06 | 5.003 | 0.965 | 52.041 | 0.633 | 0.43 | 0.18 | |
| M2UGen2024.06 | 5.104 | 0.997 | 52.246 | 0.608 | 0.433 | 0.181 | |
| VM-NET2024.06 | 6.252 | 0.899 | 67.48 | 0.986 | 0.383 | 0.147 | |
| CMT2024.06 | 8.637 | 1.22 | 85.704 | 0.08 | 0.07 | 0.124 | |
| Video2Music2024.06 | 18.722 | 1.782 | 144.881 | 0.103 | 0.023 | 0.136 | |
| MuMuLLaMAModel Type=Unified Models2026.04 | 52.25 | — | — | — | — | — |